Classification of Bengali News Articles Using Voting Classification Approach

Text classification is a crucial technology that helps in extracting and organizing textual data by automatically identifying its type based on the content. It can be applied to categorize Bengali news content into various categories. In this re-search paper, we employed the voting classification technique, which includes multiple classifiers such as Logistic Regression, Support Vector Classification (SVC), Multinomial Naive Bayes (NB), Bernoulli NB, Stochastic Gradient De-scent (SGD) Classifier, AdaBoost Classifier, Decision Tree Classifier, Logistic RegressionCV, Calibrated ClassifierCV, and Random Forest Classifier. The model was trained and tested on a gold standard dataset consisting of Bengali news content, and categorized the data into eight distinct categories, including Economy, Education, Entertainment, International, National, Science and Tech-nology, Sports, and Politics. Our results indicate that the hard voting approach achieved an accuracy rate of 89%, while the soft voting approach achieved an accuracy rate of 93%. This paper demonstrates the effectiveness of employing text classification techniques for categorizing Bengali news content and highlights the potential for its application in other languages and domains.